WeChat Article Tracker

This Python project connects to your WeChat MCP server via SSE to continuously fetch the latest articles for a list of official accounts and automatically archives them as PDFs.

Setup

1. Make sure you have Python 3.10+ installed.
2. Install dependencies:

   pip install -r requirements.txt
   playwright install --with-deps chromium

3. Edit data/accounts.txt with your list of official accounts (one per line).
4. Configure your .env file if necessary (see configuration section below).

Usage

Run the tracking service:

python -m src.main

• On the first run, if not logged in, the MCP server will generate a qrcode.png file in the current directory. Scan it with WeChat to log in.
• The service continuously runs in the background and collects new articles at set intervals.
• New articles are appended to data/articles_history.json.
• Daily snapshot reports are saved to data/daily_reports.
• PDF copies of the articles are automatically exported to data/pdf_exports.

Configuration

You can customize the behavior by setting environment variables in a .env file or directly in your environment. The system will use sensible defaults if not set.

• MCP_SERVER_URL: The URL of your WeChat MCP server (default: https://wechat-mcp.f.rwecho.top/sse).
• WECHAT_DATA_DIR: Base directory for data storage (default: data).
• FETCH_INTERVAL_HOURS: Interval in hours between fetch cycles (default: 1).
• PDF_EXPORT_ENABLED: Set to false or off to disable PDF archiving (default: true).
• PLAYWRIGHT_WS_ENDPOINT: Optional WebSocket endpoint for a remote Playwright browser instance, useful for stealth or remote execution.

Scheduler Mechanism

The scheduler automatically executes two main tasks:

1. Hourly fetches of the WeChat official accounts data without manual CLI inputs.
2. Parallel PDF printing whenever a new article is discovered to archive a permanent copy.

For Docker usage, you can simply load the application which maps the necessary persistent volumes like /app/pdf_exports and /app/daily_reports.